Ethernet is a communication standard that provides compatibility with several different data rate sub-standards that are suitable for different network bandwidths. Each generation of Ethernet typically provides a default data rate that is many orders of magnitude higher than a prior generation. The supporting communications infrastructure, such as cable and/or other media to transport signals, often lags behind the new generation, and may not be able to support the higher data rates. Reverse compatibility with older generation data rates is thus important.
While legacy compatibility for Ethernet enables new generation hardware to function with older infrastructure, the gap between the legacy data rates and new generation default data rates may be significant. For example, Ethernet transceivers that have the capability to operate at 10 Gb/s over four twisted pair channels may need to operate at one-tenth that rate, to 1 Gb/s, as the next highest supported data rate. Proposals in the art provide for sub-rates other than legacy default sub-rates.
For some Ethernet applications, operating with less than the default number of four twisted pair channels may be advantageous. A transceiver may have to be backwards compatible to existing Ethernet standards which operate on 4 pairs. Therefore, to support operation both on traditional 4-pair systems as well as reduced-pair systems, there is a need to identify whether the channel consists of 4 or fewer pairs. If this information is known in advance, it may be part of the static configuration of the system. However, if this information is not available in advance, then it will be necessary to detect the “active pairs” automatically.